PROJECT SUMMARY/ABSTRACT The proven efficacy of BRAF and MEK inhibitors (BRAFi+MEKi) for the treatment of patients with BRAF mutant metastatic melanoma testifies to a strong degree of MAPK pathway addiction. Beyond the approximately 50% of metastatic melanoma with V600BRAF signature mutations, additional subsets such as the 20% with signature mutations in NRAS may also display variable degrees of MAPK pathway addiction. However, resistance to BRAFi+MEKi therapy is commonplace, severely limiting its curative potential. The mechanisms of intrinsic, adaptive and acquired resistance are inter-related and involve selection for genomic variants, tumor cell-intrinsic adaptations, as well as stromal and immune alterations. Based on specific mechanisms of resistance, clinical trials have built on the foundation of BRAFi+MEKi and are testing ternary combinations targeting additional molecules such as AKT, c-MET and CDK4/6. Addiction to the MAPK pathway in therapy-nave melanoma can paradoxically become addiction to the MAPK pathway inhibitors (MAPKi) in therapy-resistant melanoma. We have demonstrated this for BRAF mutant melanoma (resistant to BRAFi+MEKi) and NRAS mutant melanoma (resistant to MEKi). The common basis of MAPKi addiction in resistant melanoma lies in ERK hyper-activation upon MAPKi withdrawal, regardless of the specific mechanism of resistance. Through our work in a panel of MAPKi-resistant cell lines, we have dissected the pathways downstream of ERK hyper-activation (induced by MAPKi withdrawal) that mediate either weak (a slow-cycling state) or strong (cell death) MAPKi addiction. In addition, we have identified i) a strategy to convert weak into strong addiction among chronically selected resistant populations (i.e., R-lines) and ii) a molecularly distinct R-line precursor sub-population that arises early during MAPKi treatment and displays strong MAPKi addiction. Importantly, identifying microscopic emergence of MAPKi-addicted, proliferative resistance enabled a more precise timing of drug holiday or intermittent dosing early during the treatment course to suppress resistance development. Here, we set out to test strategies to i) augment drug addiction in vivo using patient-derived xenograft (PDX) models of melanoma with acquired or late resistance to BRAFi+MEKi (BRAF mutant) or MEKi (NRAS mutant) and ii) select against microscopic resistance during the early course of MAPKi therapy in PDX models by signature-guided implementation of intermittent MAPKi dosing. This set of studies will also generate a ?renewable? tissue resource for translational melanoma work encompassing omically and clinically (in mice and men) annotated PDX models representing about 70% of metastatic cutaneous melanoma. In the era of targeted cancer therapeutics, exploiting the recently appreciated drug addiction phenotype as a common- denominator vulnerability of resistant tumors may serve to turn the inevitability of resistance against itself.